中国股市的跳跃性与杠杆效应——基于已实现极差方差的研究  

A Study on Jumps and Leverage Effect of China's Stock Markets with Realized Range-based Variance

作　　者：赵华[1] 

机构地区：[1]厦门大学经济学院,福建厦门361005

出　　处：《金融研究》2012年第11期179-192,共14页Journal of Financial Research

基　　金：国家社科基金项目(11CJY096);中央高校基本科研业务费专项基金项目(2010221055)的资助

摘　　要：以连续时间跳跃扩散理论为基础,本文将已实现极差方差分解为连续和跳跃成分,进而构建包含连续和跳跃成分的杠杆异质性自回归(LHAR-RRV-CJ)模型,对中国股市的跳跃性以及杠杆效应进行了实证研究,结果表明,中国股市具有显著的跳跃性,并存在杠杆效应,LHAR-RRV-CJ模型具有较好的预测能力。研究还发现,跳跃对中国股市波动存在显著的正的影响,其中跳跃对中国股市的短期波动影响较大,对长期股市波动影响较小。并且,杠杆效应对短期股市波动影响较大,但不影响长期股市波动。Based on the continuous - time jump diffusion theory, the paper decomposes the realized range - based variance into continuous and jump components, constructs the leverage heterogeneous autoregressive mod- el including continuous and jump components as explanatory variables, and empirically studies the jumps and leverage effect of China＇s stock markets. The results show that China＇s stock markets have significant jumps, that there is leverage effect in China＇s stock markets, and that LHAR - RRV - CJ model has better predictive power. It is found that the jumps have a significant and positive impact on China＇s stock market volatility, espe- cially having greater influence on the short -term volatility and less influence on the long -term volatility. Mo- reover, leverage effect has a greater impact on the short - term stock market volatility, but not impacting on the long - term volatility.

关 键 词：已实现极差方差 跳跃 LHAR—RRV—CJ模型 杠杆效应 

分 类 号：F224[经济管理—国民经济] F832.51